University Events

Description: TOPIC: APPLYING ARTIFICIAL INTELLIGENCE TO SECURE CONNECTED VEHICLES

LOCATION: Lester W. Cory Conference Room, Science & Engineering Building (Group II), Room 213A

ABSTRACT:
Vehicles, equipped with sensing, communication, processing and storage devices, form a transportation ecosystem, which enhance driver safety with collision avoidance, raise transportation efficiency by traffic forecast, and bring in convenience and entertainment via real time travel planning and tourist advertisement. Connected vehicles "talk" to each other and, thus, form vehicular ad hoc networks (VANET). VANET inherits vulnerability from wireless connection or internet access. Therefore, it is critical to protect connected vehicles against hacking and security breaches. The traditional encryption and authentication methods are not suitable to VANET due to high speed mobility and ad hoc nature. Most research activities study either safety with laboratory experiments or efficiency with VANET simulators. Few researches have been conducted on the security and safety protection for VANET.

This project explores the application of artificial intelligence to secure connected vehicles. Artificial Intelligence (AI) is an area of computing to simulate intelligent behavior in problem solving. By adopting context-adaptive signature verification strategy, similar to human comprehension, the work reduces the computational overhead in validating beacon messages for secured VANET communications. The context-adaptive signature verification strategy requires checking the signature of the initial beacon of new vehicles and from then on it checks for every nth beacon. To prevent intermediate spoofed beacons, a linear estimation methodology is used to estimate future position and speed of the neighboring vehicle. If the estimated positions and those reported in the beacons differ greatly, then signature verification is triggered. Our work starts with understanding of the interplay between vehicle traffic and data traffic in VANET through VANET simulator. We then model the new security scheme with MATLAB. The model indicates the significant saving in computational overhead of our scheme comparing to conventional methods of message verification.

Future work will test the scheme under various scenarios. We will also improve the scheme by reducing communication overhead.

NOTE: All ECE Graduate Students are ENCOURAGED to attend.
All interested parties are invited to attend. Open to the public.

Advisor: Dr. Hong Liu
Committee Members: Dr. Liudong Xing and Dr. Honggang Wang, Department of Electrical & Computer Engineering

*For further information, please contact Dr. Hong Liu at 508.999.8514, or by via email at hliu@umassd.edu.